AIM: To observe the biologic behavior of pancreatic cancer cells In vitro and in vivo, and to explore the potential value of angiostatin gene therapy for pancreatic cancer. METHODS: The recombinant vector pcDNA3.1(+)-anglostatin was transfected into human pancreatic cancer cells PC-3 with Llpofectamine 2000, and paralleled with the vector and mock control. Angiostatin transcription and protein expression were determined by Immunofluorescence and Western blot. The stable cell line was selected by G418. The supernatant was collected to treat endothelial cells. Cell proliferation and growth in vitro were observed under microscope. Cell growth curves were plotted. The troms-fected or untroms-fected cells overexpressing angiostatin vector were Implanted subcutaneously into nude mice. The size of tumors was measured, and microvessel density count (MVD) in tumor tissues was assessed by Immunohlstochemistry with primary anti-CD34 antibody. RESULTS: After transfected into PC-3 with Lipofectamine 2000 and selected by G418, macroscopic resistant cell clones were formed In the experimental group transfected with pcDNA 3.1(+)-angiostatln and vector control. But untreated cells died in the mock control. Angiostatin protein expression was detected in the experimental group by Immunofluorescence and Western-blot. Cell proliferation and growth in vitro in the three groups were observed respectively under microscope. After treatment with supernatant, significant differences were observed in endothelial cell (ECV-304) growth in vitro. The cell proliferation and growth were Inhibited. In nude mice model, markedly inhibited tumorlgenesls and slowed tumor expansion were observed In the experimental group as compared to controls, which was parallel to the decreased microvessel density in and around tumor tissue. CONCLUSION: Angiostatin does not directly inhibit human pancreatic cancer cell proliferation and growth in vitro, but it inhibits endothelial cell growth in vitro. It exerts the anti-tumor functions through antiangiogenesis in a paracrine way in vivo. AIM: To observe the biologic behavior of pancreatic cancer cells In vitro and in vivo, and to explore the potential value of angiostatin gene therapy for pancreatic cancer. METHODS: The recombinant vector pcDNA3.1 (+) - anglostatin was transfected into human pancreatic cancer Cells PC-3 with Llpofectamine 2000, and paralleled with the vector and mock control. Angiostatin transcription and protein expression were determined by Immunofluorescence and Western blot. The stable cell line was selected by G418. The supernatant was collected to treat endothelial cells. and growth in vitro were observed under microscope. Cell growth curves were plotted. The troms-fected or untroms-fected cells overexpressing angiostatin vector were Implanted subcutaneously into nude mice. The size of tumors was measured, and microvessel density count (MVD) in tumor tissues was assessed by Immunohlstochemistry with primary anti-CD34 antibody. RESULTS: After transfected into PC-3 with Lipofectamin e 2000 and selected by G418, macroscopic resistant cell clones were formed In the experimental group transfected with pcDNA 3.1 (+) - angiostatln and vector control. But untreated cells died in the mock control. Angiostatin protein expression was detected in the experimental group by Immunofluorescence After treatment with supernatant, significant differences were observed in endothelial cells (ECV-304) growth in vitro. The cell proliferation and growth were Inhibited. In nude mice model, markedly inhibited tumor cells and slowed tumor expansion were observed in the experimental group as compared to controls, which was parallel to the decreased microvessel density in and around tumor tissue. CONCLUSION: Angiostatin does not directly inhibit human pancreatic cancer cell proliferation and growth in vitro, but it inhibits endothelial cell growth in vitro. It exerts the anti-tumor functions through antiangiogenesis in a paracrine way in vivo.